Membrane based electrochemical cells, and particularly, proton exchange membrane (PEM) fuel cells are well known. PEM fuel cells convert chemical energy to electrical power with virtually no environmental emissions and differ from a battery in that energy is not stored, but derived from supplied fuel. Therefore, a fuel cell is not tied to a charge/discharge cycle and can maintain a specific power output as long as fuel is continuously supplied. Significant funds have been invested in fuel cell research and commercialization, indicating that the technology has considerable potential in the marketplace. However, the high cost of fuel cells when compared to conventional power generation technology deters their widespread use. The cost of fabricating and assembling fuel cells can be significant, due to the materials and labor involved. Indeed, as much as 85% of a fuel cell's cost can be attributed to manufacturing.
Traditionally, one of the problems of using internally manifolded stacks in fuel cells and other electrochemical applications, is the area that is sacrificed in sealing around the internal manifolds. One remedy is to locate some or all of the manifolds external to the stack. Several examples exist but all suffer from difficulties in sealing between the manifold and the stack. As in traditional stacks, sealing is typically accomplished with gaskets and compression. Unfortunately, gasket/compression based seals have a number of inherent drawbacks, including a sensitivity to thermal cycling, requirements of uniform compression and associated hardware, high tolerance parts, and delicate assembly requirements.
In general, a single cell PEM fuel cell consists of an anode and a cathode compartment separated by a thin, ionically conducting membrane. This catalyzed membrane, with or without gas diffusion layers, is often referred to as a membrane electrode assembly (“MEA”). Energy conversion begins when the reactants, reductants and oxidants, are supplied to the anode and cathode compartments, respectively, of the PEM fuel cell. Oxidants include pure oxygen, oxygen-containing gases, such as air, and halogens, such as chlorine. Reductants, also referred to herein as fuel, include hydrogen, natural gas, methane, ethane, propane, butane, formaldehyde, methanol, ethanol, alcohol blends and other hydrogen rich organics. At the anode, the reductant is oxidized to produce protons, which migrate across the membrane to the cathode. At the cathode, the protons react with the oxidant. The overall electrochemical redox (reduction/oxidation) reaction is spontaneous, and energy is released. Throughout this reaction, the PEM serves to prevent the reductant and oxidant from mixing and to allow ionic transport to occur.
Current state of the art fuel cell designs comprise more than a single cell, and in fact, generally combine several MEAs, flow fields and separator plates in a series to form a fuel cell “stack”; thereby providing higher voltages and the significant power outputs needed for most commercial applications. Flow fields allow for the distribution of the reactants through the fuel cell and are typically separate from the porous electrode layers within the fuel cell. Depending on stack configuration, one or more separator plates may be utilized as part of the stack design to prevent mixing of the fuel, oxidant and cooling streams within the fuel cell stack. Such separator plates can also provide structural support to the stack.
Bipolar plates perform the same function as an oxidant flow field, fuel flow field and separator plate in combination and are often used in the design of fuel cells as their use can reduce the number of components required in the functioning fuel cell. These bipolar plates contain an array of channels formed in the surface of the plate contacting an MEA which function as the flow fields. The lands conduct current from the electrodes while the channels between the lands serve to distribute the reactants utilized by the fuel cell and facilitate removal of reaction by-products, such as water. Fuel is distributed from the fuel inlet port to the fuel outlet port, as directed by the channels, on one face of the bipolar plate, while oxidant is distributed from the oxidant inlet port to the oxidant outlet port, as directed by the channels, on the opposing face of the bipolar plate, and the two faces are not connected through the plate. The particular design of the bipolar plate flow field channels may be optimized for the operational parameters of the fuel cell stack, such as temperature, power output, and gas humidification. Ideal bipolar plates for use in fuel cell stacks are thin, lightweight, durable, highly conductive, corrosion resistant structures such as carbon/polymer composites or graphite. In the fuel cell stack, each bipolar plate serves to distribute fuel to one MEA of the stack through its fuel flow field face while distributing oxidant to a second MEA through the its opposite oxidant flow field face. A thin sheet of porous paper, cloth or felt, usually made from graphite or carbon, may be positioned between each of the flow fields and the catalyzed faces of the MEA to support the MEA where it confronts grooves in the flow field to conduct current to the adjacent lands, and to aid in distributing reactants to the MEA. This thin sheet is normally termed a gas diffusion layer (“GDL”), and can be incorporated as part of the MEA.
Of necessity, certain stack components, such as the GDL portion of the MEA, are porous in order to provide for the distribution of reactants and byproducts into, out of, and within the fuel cell stack. Due to the porosity of elements within the stack, a means to prevent leakage of any liquid or gases between stack components (or outside of the stack) as well as to prevent drying out of these porous elements due to exposure to the environment is also needed. To this end, gaskets or other seals are usually provided between the surfaces of the MEA and other stack components and on portions of the stack periphery. These sealing means, whether composed of elastomeric or adhesive materials, are generally placed upon, fitted, formed or directly applied to the particular surfaces being sealed. These processes are labor intensive and not conducive to high volume manufacturing, thereby adding to the high cost of fuel cells. Additionally, the variability of these processes results in poor manufacturing yield and poor device reliability.
Fuel cell stacks may also contain humidification channels within one or more of the coolant flow fields. These humidification channels provide a mechanism to humidify fuel and oxidants at a temperature as close as possible to the operating temperature of the fuel cell. This helps to prevent dehydration of the PEM as a high temperature differential between the gases entering the fuel cell and the temperature of the PEM causes water vapor to be transferred from the PEM to the fuel and oxidant streams.
Fuel cell stacks range in design depending upon power output, cooling, and other technical requirements, but may utilize a multitude of MEAs, seals, flow fields and separator plates, in intricate assemblies that result in manufacturing difficulties and further increased fuel cell costs. These multitudes of individual components are typically assembled into one sole complex unit. The fuel cell stack is formed by compressing the unit, generally through the use of end plates and bolts although banding or other methods may be used, such that the gaskets seal and the stack components are held tightly together to maintain electrical contact there between. These conventional means of applying compression add even more components and complexity to the stack and pose additional sealing requirements.
Various attempts have been made in the fuel cell art to address these deficiencies in fuel cell stack assembly design and thereby lower manufacturing costs. However, most still require manual alignment of the components, active placement of the sealing means and/or a multi-step process, each of which presents notable disadvantages in practice. See, e.g., the processes described in U.S. Pat. No. 6,080,503, to Schmid et al., U.S. Pat. No. 4,397,917, to Chi et al., and U.S. Pat. No. 5,176,966, to Epp et al.
Additionally, in traditional fuel cell cassettes, two types of MEAs dominate; MEAs in which 1) the membrane extends beyond the borders of the gas diffusion layers, and 2) gasket materials are formed into the edges of the MEA itself with the membrane and GDLs approximately of the same size and shape (see, e.g., U.S. Pat. No. 6,423,439 to Ballard). In the first type, separate gaskets are used to seal between the membrane edge extending beyond the GDL and the other part of the stack (bipolar plates). In the second type, the gasket of the MEA seals directly to the other parts of the stack. Each of these methods requires compression to make a seal. These compressive-based seals require that all the components in the stack have high precision such that a uniform load is maintained. MEA suppliers have become accustomed to supplying the MEA formats above.
Still other attempts have been made to improve upon fuel cell design and performance. For instance, U.S. Pat. No. 4,212,929 describes an improved sealing method for fuel cell stacks. That patent reports a sealing system which utilizes a polymer seal frame clamped between the manifold and the stack. As described, the seal frame moves with the stack and the leak rate associated with a typical manifold seal is reduced during compression. U.S. Pat. No. 5,514,487 and U.S. Pat. No. 5,750,281 both describe an edge manifold assembly which comprises a number of manifold plates. The plates are mounted on opposite sides of the fuel cell stack and function in such a way to selectively direct the reactant and coolant streams along the perimeter of the stack. While these designs offer limited improvements to other conventional assemblies, they are generally unsuitable for high-volume manufacture.
Recognizing these and other deficiencies in the art, we have developed a series of innovative methods for sealing manifold ports within the stack or a module thereof, as well as methods for sealing the stack or module periphery that are less labor intensive and more suitable to high-volume manufacturing processes (see World Publication WO 03/036747, incorporated herein by reference). That publication discloses a ‘one-shot’ assembly of fuel cell stacks (and other electrochemical devices) in which all of the component parts are assembled into a mold without gaskets. A resin is introduced into the mold and this resin selectively penetrates certain portions of the assembly either by resin transfer molding or injection molding techniques. Upon hardening, that resin seals the components and defines all the manifold channels within the stack. The net effect is to replace the gaskets of the traditional stack with adhesive based seals, introduced after the assembly of the components.
We also have previously described fuel cells having an MEA in which the GDL and membrane were more or less of the same general outline as each other and of the overall stack profile (see World Publication WO 03/092096 A2, incorporated herein by reference). The major advantage of this technique is the ability to directly use a roll-to-roll MEA without having to do any post processing. However, a substantial portion of the cross-section of each MEA is used for sealing the various manifold openings and periphery of the stack such that only about 50% of the cell cross section is used for the electrochemical reaction.
We also have developed membrane-based electrochemical cells, and more particularly, PEM fuel cell stacks which comprise one or more composite MEAs having a molded gasket about the periphery. The gasket portion of the composite MEA has one or more features capable of regulating the flow of sealant during sealing processes (see International Application PCT/US03/37127, incorporated herein by reference).
In another previous patent application, we have reported on an innovative fuel cell stack design which assembles together individual modules to form a fuel cell stack of requisite power output where each module permanently binds a number of unit cells together (see World Publication WO 02/43173, incorporated herein by reference).
Despite even our own advancements over the prior the art, we have recognized that further improvements can be made to the technology. One improvement, for example, would be to utilize a more significant portion of the total MEA area for the electrochemical process. For instance, with particular reference to those fuel cell stacks which include an internal manifold design, a certain cross-section of the cassette must be utilized for sealant channels and reactant/coolant manifolds; thus, that potentially active area is necessarily sacrificed. It also would be desirable to provide an improved fuel cell stack design that is less complex, more reliable, and less costly to manufacture. Additionally, it would be highly desirable to provide improved fuel cell stacks having reduced weight and size and (as noted above) in which a greater percent of the total MEA surface area is available for use in the electrochemical reaction occurring within the stack, e.g., available for catalyst area and proton transfer.